View Single Post
  #6   Report Post  
Posted to rec.audio.tubes
Wessel Dirksen Wessel Dirksen is offline
external usenet poster
 
Posts: 11
Default Recommendatons bias settings for KT88 and KT90

On 11 sep, 16:44, Patrick Turner wrote:
Wessel Dirksen wrote:

Hi Guys,


I've been droping in again as I have some time and ideas. Even if my
amps sound good, after awhile you still want to do something different
and start taking them apart again. LOL


I was hoping to get some opinions about the "sweet spot" for KT88 and
KT90's. I bought a set of both. PP 4x power tubes at AB1 with B+ 425V.
I want to run a bit rich on class A if possible. What sounds best with
a descent life span?


BTW, I also got some NOS EI 6CG7's. Wow, what a diference a tbe makes!


Thanks,


Wessel


If you want lots of class A and blameless fidelity, you must use at
least 40% UL,
or triode, and bias the tubes up a bit, and keep Ea low,
and RLa-a high, so this takes care of the output stage.

But didn't tell us what schematic you have, so whatever we might say may
be of zero benefit to your music.
If you have cathode biasing, Ek with UL/triode will be about 45V, so Ea
= 380V approx,
so adjust Rk to get Ik = about 80mA, for Pda = 32 watts.


Yeah, forgot to mention details. I really like the triode sound (+/-
30 watts is enough for me) and use cathode bias.
I was already thinking of about 80mA as a target. Healthy amount of
Class A with what I guess in can infer from your input that this
won''t have too much wear and tear.

Then you might get up to 28 watts of pure class A and maybe 10 more
watts of AB in UL.
Less for triode, about 27 watts.

Loading is important, and expect Va swing at each anode = +/- 300V pk in
UL,
and if the Ichange = +/- 75mA pk in class A, the load each tube works =
4k, so
load a-a = 8k and PO = 22 watts at low thd and will all be class A.
Lower loads such as 4ka-a will be tolerated OK and give more power, AB.

But if you have a 8ohm and 4 ohm outlets on the back of the amp, try the
4 ohms regardless of what your speakers are.
This will reflect the highest possible load to the tubes depending on
the OPT TR.
The highest load for UL or triodes give the lowest THD/IMD because the
high RL gives lowest THD/IMD, and the higher gain, therefore increasing
applied global NFB.
But about 12dB of NFB with triode is all you need, maybe 15dB with UL,
so perhaps you may adjust NFB downwards after maximizing output tube
gain with class A action into high RL
when Ia is high and gm is thus high, and Ra low.


I'm only using about 5 db of global NFB. Great soundstage!

NOS Seimans 6CG7 are supposed to be the best sounding, but for blameles
dynamics,
use EL84 in triode each side of the LTP driver with 15mA per tube.


Hummm, interresting. EL84 better than 6CG7 as a driver? Robust yes,
but also better sounding?

FWIW, the current config (experiment) is a two input stage amp with 2x
6CG7's
A CCS non bypassed grounded cathode stage biased at about 6.7mA is DC
coupled (+ voltage PS only) to a long tail phase splitter (& driver)
biased at about 14,5 mA. I'm using different Rl values to offset the
balance problem. It's cool because there is only one capacitance in
series with the signal into the power stage (f-3 about 15 hz). Both
stages are good for a gain of 24. With 2VRMS input it seems to drive
it all fine. OPT is an older torroidal Plintron with Raa of 2400.
Currently I'm using the Blumlein "garter" method to crossbias the big
tubes. It takes a 20V B+ hit but distortion is lower.

I don't have the most eloquent measuring gear for beyond 96khz. But
it's very linear with a speaker load (+/- 1.5 db) and overall THD and
IMD are about 1% at 3/4 full output with EL34's running at 65mA. I'm
kinda proud of this first contraption of mine in it's current state.

And choke loading if you can use it; there is info on this at my site,
see the pages on 8585 amp.
6CG7 are very good in the input stages, like 6SN7.

Patrick Turner.- Tekst uit oorspronkelijk bericht niet weergeven -


Thanks Patrick